I'm afraid I don't understand the question. I guess you are referring to this PR?
It seems wrong in multiple ways (for starters, they change the module path), but it also seems outdated; you have put several commits on master since then.
From a casual glance at your code there is at least one race-condition: While SetLang is protected by a mutex, none of the other methods of Printer are. So calling SetLang concurrently with any other method will create a data-race.
Apart from that the PR changes the API quite a bit and the "memory leak" they talk about seems to be something they introduce themselves in the first commit by using a global map and not provide a way to remove an element from it - and then they fix that in the second commit by adding a new function.
Overall, I'm afraid to provide help, there needs to be a clear question though. If you are indeed wondering about the PR, I think the best way to address that would be to ask the author; it's their change and it's fine to ask for clarification of what they are trying to achieve before accepting it.